Kirk - Three Early Saxon Graves From Dorchester, Oxon

7/29/2019 Kirk - Three Early Saxon Graves From Dorchester, Oxon

1/16

Three Early Saxon Graves from Dorchester, Oxon.By jOA."I R. KIRK and E. T. LEEDS

T H E environs of the modern village of Dorchester, which lies about eightmiles south of OKford at the confluence of the rivers Thame and Thames,have yielded three important inhumation-burials, which, from their accompanying grave-goods, clearly belong to the late fourth or earliest fifth centuryA.D . , and represent some of the earliest Germanic people to come to this countryat the end of the Roman occupation.one of these graves is a new discovery. The first two have, since theycame to light in 1874, been repeatedly cited as providing some of the earliestmaterial evidence for the presence of the Saxons in this country, but no accountwith a complete catalogue of the relics preserved has ever been published. Afresh survey has suggested that such a catalogue is required, if they are to beproperly understood and their interest fully appreciated, the more so as the newsurvey has thrown some light on the purpose of the contents of the principalgrave (Grave I ). These hitherto have appeared to comprise a mass of disjectamembra for which no obvious explanation had been offered ..The first two graves (Ashmolean Museum, 1886. (448) from which thisarchaeological material came, according to information given to E.T.L. inDecember 1914 by Mr. A. H. Cocks of Yewdon, near Henley,' were found inthe top of the Dyke Hill, which, according to the Rev. W. C. MacFarlane,vicar of Dorchester, was ' the end barrow nearest the Thame stream at theDyke Hills'. They came to light in the course of levelling the eastern end ofthe Dyke Hills, and the objects found passed into the possession of the Depart-ment of Comparative Anatomy, whence they were transferred to the Ashmoleanin 1886. They were not the only burials discovered there, since, among theRolleston MSS. in the Department of Comparative Anatomy are' Notes on theA. H. Cocks Collection from Dorchester Dykes', dated between 4 March and24 April 1871, and mentioning other inhumations. Two of these were foundat depths of 4 and 8 ft. in gravel-silt, the second of them in a sitting posture-apparently a Bronze Age burial like those from Cassington and Eynsham. Allothers (number not stated) were found in vegetable soi l ' and ' very shallow'.

, Mr. Coc.k5 was an undergraduate at the date of the demolition, and not only wu actively con-cerned in the recovery of the relics, but was also instrumental in securing the remainder of the dykesfrom destruction.


2/16

: II

FIG. 27DORCHESTER, OXON.1-13. object! from Grav(' I, (pp. 65 fr., 70 fr.)14-16, brooches and bucld(' from Crave 11 .PP 67 fr., 72 f.)

Scale' t64

\


3/16

THREE EARLY SAXO GRAVES FROM DORCHESTER, OXON.Except for the presence of coffin-nails with a powerful man' in the 4 ft. deepburial, no relics were available for determining the date.

The third grave dealt with in the present article was found in the Minchinrecreation ground, at the northern end of the village, immediately east of themain road to Oxford. Its furniture was originally acquired for the Pitt RiversMuseum in 1914, whence the relics were transferred by exchange to theAshmolean Museum in 1940.

GRAVL I

[ J { W I I I I U ~ U \ l l I l I U0

:.J 010 0I 0[ 0 0

0

I{. 00 0vI, , ,

flo FIC. 28

DORCHESTER. OXON.

0

0)

01ot0)000vo}, ," . .' ,.1.".l0

S u ~ g e s t e d reconstruction of apron from Grave I (p. 71 \

THE CONTENTS OF THE GRAVESThe Rolleston notes dated 29.3.74 supply particulars of this grave.

Orientated' SSE.-I\'N\\,., head at SE. end, knees apparently bent; skeleton ofenormous size', estimated in a later note, from a Libia 16 in. long, as 6 flo Therelics recovered from the grave were:

(a) Buckle, bronze, L. 2-t. in., W. 2 , ' . in . found by shoulder' (PL. IV, A,FIG. 27,1)'; the ring, which shows considerable signs of wear, is D-shaped,terminating in heads of animals with open jaws; the tab is semicircular, with

1 Thi's and all the other drawings which illustrate this article are by Mrs. E. Cox of the Ash-molean Museum: the photographs were taken in UlC Ashmolean Museum studio.


4/16

JOAN R. KIRK AND E. T. LEEDSstamped circles and incised lines round the edge, and three rivets still inposition; it has been broken and mended in antiquity, a new piece havingbeen added between the tongue and the right end of the ring. On the tongueare two horizontal incised lines.(b) Strap-tag, bronze, L. 2 in., W. [n in. (PL. [V, A, FIG. 27, [[); the hutt [ssuh-triangular, with two notches on the top edge, which is slightly hevelled,and has two rivets still in position; the bu tt is divided by two incised lines fromthe pear-shaped end.(c) Three disk-ended attachments, bronze, , found by the thighs' (PL. [V, A,FlG. 27,5-10"), each consisting of: (i) a central ring, quoit-shaped, with a lip atthe inner edge on one side, D. n n., { . in. and H- in. (ii) a flat piece of bronzelooped round the ring; only two of these remain; the front of one of them isdecorated with groups of incised lines, between which the edges are bevelled;the end is notched; the front of the other is decorated with four notches at theedge, followed by one single and two sets of double transverse lines. There aretwo rivets attached to the underside of the front, which do not pass through theback:' L. [l.in., th in . (iii) a flat disk, decorated with a scalloped edge andconcentric circles, two of which are corded. Each had had a loop, whichpassed through the ring (i), doubled back and was attached to the disk by arivet, which was driven through the centre. The fronts of the loops aredecorated with two corded horizontal lines and two diagonal lines. Thesmallest disk is very much worn, and has lost most of its loop. D . in., [t'o in.,[,'0- in.

(d ) Garniture


5/16

THREE EARLY SAXON GRAVES FROM DORCHESTER, OXON.each rivet-hole, the notches beneath the central hole being spread outin a fan-like motif.

For a suggested reconstruction of the object to which these pieces belonged seeFIG. 28 and p. 71 below.(e) Two bars of exactly similar design and decoration to those describedunder (d) (ii) above,but only 2-? in. long and h in. wide (PL.IV,A,FIG. 27,4);each has lost one end rivet, and the central rivet of one still has a square piece ofhronze attached to the end.( j) Disk or bead, bone, D. I-r'o in. , T. T" in. (PL. IV, A, FIG. 27, 12), decoratedon top with six double-ringed bull's-eyes circles alternating with six pairs ofsmall single-ring bull's eyes, and round the edge with six double-ringedalternating with six single-ringed.

(g) Double plate of hin bronze, L. 2 in., W. -r'o in., originally riveted togetherat one end, so as to swivel ( PL. IV, A, FIG. 27, 13); the upper plate is oblong,rounded at the free end, and with a rectangular notch Iii in. wide, , . . in. longcut into the edge towards the riveted end; the lower plate is rectangular, withtwo rivet-holes at the free end (one rivet remaining) ; down the centre is anarrow slit T'o in. wide, widening out into a circle 130 in. in diameter at the freeend. I t appears that some object could have been inserted in this circular hole,slid down the slit, and clipped firm by the notch in the upper plate. On adiminutive scale the whole object closely resembles an army buttonstick.

(h) Thin platr, bronze, L. -h in., circular at one end D. , ' . in., and rectangular at the other, with a rivet at the rectangular end ( PL . IV , A, FIG. 27,2).

From notes in the Rolleston MS. the grave seems also to have containedweapons. The Rev. W. C. Macfarlane, in a letter dated 23 March 1874,writes 'several pieces of iron, I believe, were thrown into the river J, andRolleston hirnselflists' pieces of iron ' and adds' I should ascribe to the man aniron knife 5,'. in. long.GRAVE 11

Skeleton of moderate size', of' a woman, the skull showing her to havebeen big.' The recorded contents of the grave are:(a) Brooch, bronze, L. 2-r'o in., of early cruciform type, with faceted footwith no animal decoration ( PL . IV, B, FIG. 27, 14) ; a ridge running down thecentre or the bow, a knob at the back of the head and a spring for the pin.(b) Back-plate, bronze, of an applied brooch (Roeder's komponierte Fibel,cf. those found in Grave III ), D. Ii. in. (PL. IV, B, FIG. 27, IS) , with two rectangular slots for insertion oflhe catch-plate and spring; on the upper side canbe seen traces of the cement with which the decorative plate was originallyfixed.


6/16

8M

)

'( ' OW I

I I !!II ittI'e?" i . db. Q-~ ; - - Q -

I1IC. 29DORCHESTER, OXON.Objf'Cts from Gran' 111 (p. 69 f. )

Scale: i68


7/16

THREE EARLY SAXO:'> GRAVES FROM DORCHESTER, aXON.(c) Buckle, bronze, L. 2T'.in., W. of plate 1'0 in. (PL. tV, B, FIG. 27,16) ; the

ring is D-shaped and has, at the top, two stylized animal heads confronting eachother with open jaws, and twO projecting horse protomi facing outwards. Theplate is rectangular and composed of two thin plates of bronze with a deepnotch at the upper end to take the tongue, the two shoulders thus formed beingriveted together to hold the ring.One has been broken and repaired inantiquity. At the other end are tworivets. The upper plate is decoratedwith a border of dots, inside which,down the two long sides, is a run -ning scroll; down the centre of theplate are three flaming circles alternating with two hatched lozenges.GRAVE III

(a) Pair of brooches, bronze, ofapplied type, D. q ~ in. (PL. v, B,FIG. 30, 1 and 3) ; one has pin andcatch complete, except for half thespring, the other has only the springholder remaining. On the lallerthe applied disk is very nearly complete, on the former nearly half ismlssmg. In both cases the outeredge has disappeared. The decoration on the applied disk consists oftwo rings of dots, within which is awhirligig design of five arms, eachending in a ball and having an open

I 3~FIG. 30

DORCHESTER, OXON.Applied brooches from Grave III (pp. Gg, 73 f.)

Scale; ttriangle at the top; in the centre of the whirligig is a plain circle, surrounding a circle of dots within which is a central boss with a spot in the middle.

(b) Back-plate, bronze, ofa similar brooch, D. I". in. (PL ., v, B, FIG. 30, 2);the catch is missing, but the holder for the hinged pin remains; through thecentre is a small hole.(c) Pair


8/16

JOAN R. KIRK AND E. T. LEEDS(e) Key, bronze, L. 2t'. in., forlever lock (PL. v, A, FIG. 29, I) ; at the top ofthe shank is a large ring, D. in., surmounted by a moulded knob.(f) Seven rings, bronze, D. varying from 2,'. in. to ! ~ in. (PL. v, A,

FIG. 29, nos. 4, 6-10, (2); the largest circular in section, the next lozengeshaped with a ridge on each side (cf. the rings on the attachments, no. (c) (i) inGrave I) ; the smallest ring but one is more similar to those from Grave I inthat it only has the ridge on one side. The remaining rings are lozengeshaped or flattened in section.(g) Fragment i f pricker, bronze, L. 24: in., with splayed and perforatedhead broken off at the perforation (PL. v, A, FIG. 29, II ) ; the point is alsobroken. Beneath the perforation are three stamped bull's-eye circlets.(h) Finger-ring, bronze, D. i?J in., flat band in three spiral coils, the endsbeing pointed (PL. v, A, FIG. 29, 14).(i) Fragment i f lat band, bronze, L. 6/. in., curved, and decorated with asuccession of bull's-eye circlets between borders of small crescents (PL. v, A,FlG. 29, (3 ).

(j) Coin ofValens (A.D. 364-378). Rev. Securitas Reipublicae (PL. v, A).Very worn.NOTES ON THE GRAVE-GOODS AND COMPARISONS

WITH OTHER FINDSGRAVE I

(a) Parallels to ille bronze buckle are figured by Pilloy, Cimetieres dans l' Ais.e,I , pI. v, from Abbeville, and u, COld. pI. 2, from Vermand. More common,possibly rather later, are buckles with rectangular plates, which also havehoops with confronted lion-heads.(b) Strap-tag, cpo Vermand, Pilloy, u, COld. pI. 3 ; Belleray, Meuse, ibid.,I, 244 fig. Examples have also been noted from Eprave, Samson, Seraing(Belgium), and in a longer form from Borgstedt and Quelkhorn ( . Germany).English parallels come from Milton-next-Sittingbourne, Kent (MaidstoneMuseum) and Croydon, Surrey (V.C.H. Surrey, I, pI. facing p. 257, fig. 7).

(c) Disk-attachments, cpo Vermand, Pilloy, fl, 235, pI. 16, 5a, 6a, 7a and26b (three examples as at Dorchester) ; Belleray, ibid., I, 243, fig. Il l , pI. IV , I I .They are known from Furfooz (J. A. E. Nenquin, La Nieropole de FurJooz, Bruges,1953, pI. XI), Herstal, Span tin and Tongres (Belgium) ; in N. Germany fromRitzebiittel, Perlberg, and also Galgenberg near Nesse (Fr. Roeder, Neue Fund.auf Kontinental-siichsischen Friedhiifen der Viilkerwanderungszeit (Anglia, Bd. 57),Tar. VI, 4, from burial 8 with cremation). In this country they have beenfound at Richborough, Kent (Bushe Fox, Richborough, IV , pI. XXXII, 7a, pI. Lll,


9/16

THREE EARLY SAXO GRAVES FROM DORCHESTER, OXON.187-8), and Croydon, Surrey (V.C.H. Surrey, I, pI. facing p. 257, fig. g). Theirpurpose is discussed by G. Baldwin Brown, Arts in Early England, IV, 558-9,where he associates them with the disk-armoured apron worn by Roman soldiers,well represented on stelae (e.g. Lindenschmit, Alterthiimer u.h. Vorzeit, I, pt.x, pI. 5). He rightly, however, observes that the disks figured on the aprons(German Riemens,hIlrzen ) are always portrayed in large numbers. In factthey are no more than plain or decorated disk-headed rivets, affixed to a seriesof pendent straps (see pieces of straps so fortified from Vermand (Pilloy, op . ,it.,I I , pI. 16, ga, 23b, and 24b)), the straps being finished off by a set of pendentornaments of various forms.' The disk-attachments here under considerationare apparently found singly, or at most in three examples as at Dorchester andat Vermand, and could not, as Pilloy maintained, be directly connected with theapron. Their purpose can be more fittingly described after discussion of thenext item, (d) .(d) Bronze oblong riveled plates. It has already been noted that all these aresaid to have been found among the ribs of the skeleton, and there can be Littledoubt that they constitute the fortifying components of an apron. Eleven ofthem are identical in length, and, when placed together with ,'0 in. spacebetween them to allow for flexibility of the leather to which they were riveted,they formed a bronze-armoured apron or sporran, approximately 5 in. long(FIG. 28). The term sporran is preferable because the plates thus arranged,that is transversely, instead of vertically, as in the normal Riemens,hurz of themonuments, would necessarily be more rigid than the pendent straps of theother type. The irregularity of the rivets suggests that the bars were attachedto a leather apron, rather than to three separate, vertical leather straps. Wehave found no parallel to the proposed reconstruction, which seems the onlypossible explanation of the material. The two short, narrower plates caneasily be brought into the make-up of such an equipment, and are here leftunexplained, though their form and decoration agrees WiIh that of the longerplates. Parallels to the upper bar with its ribbed cylindrical element arenumerous. Pilloy (op. cit., I, 240) writes of their occurrence at Samson' enassez grand quantite'. On those found at Vermand (op. ,it., I I , pI. 16, la , 2a, 2Iband 22b) Pilloy remarks' Vermand a donne quatre de ces plaques ... qui setrouvaient par paires sur Ie me me sujet '. In that case the Riemens,huTZ of themonuments could be worn both at front and back, though perhaps morenormally only in front, since one cylinder-plate seems to be usual. Severalare known also from Furfooz, Tongres and other sites. We have noted onefrom Ritzebiltte1 (Kie1 Mus. ) and in this country from Milton-next-Sitting-

.. It may be suggested that such were the objects illustrated in R ~ h h o r o / J . g h , II, pI. XXI, 47-9, andtV , pl. LVI, 263.

71


10/16

JOAN R. KIRK AND E. T. LEEDSbourne (Maidstone Mus.), and Richborough (C. Roach Smith, AntiquitiesofReculver, pI. v, 2 ) .

Here we can return to the disk-attachments, examples of which have alsobeen found at Richborough. There can hardly be any doubt that they wereused on baldrics or shoulder-straps, as portrayed, usually singly, on the stelae.That more than one could be necessary seems certain. No bell unsupportedby shoulder-straps could have borne the weight of sword, dagger and apron;a third might be required to carry a shield.

(e) Bone bead. Rolleston, as appears from his notes, evidenUy insisted thatit was a spindle-whirl and should be assigned to Grave II ; indeed he so lists it.But in the face of the testimony, not only of the v;car of Dorchester, but of themen working on the site, it certainly came [rom Grave I. It must thus haveserved as a toggle for some purpose or other or as a sword-knot, as seems tohave been the purpose of such large beads in Anglo-Saxon graves, e.g. W. M.Wylie, Fairford Graves, pI. iv.

For the other objects figured no parallels appear to be available. Thelittle object (f) may have been affixed to the belt or even to a baldric (ba/teus) tosecure some small item of equipment, like a whistle, which it was desirable tohave easily detachable.GRAVE 11

(a) Fibula. This specimen has been widely recognized as the earliest ofits class so far found in England. The pin-catch extending the whole lengthof the foot is compared by Aberg with that from Beetgum, Friesland (AngloSaxons in England, fig. 43), bu t in fact it represents an earlier stage, sinceit bears no trace, as does the Beetgum brooch, of any development o[the animalhead terminal, which later becomes a most marked feature. The date assignedto it lies in the last quarter of the fourth century or around 400 (Salin, Manadsblad, ]894, p. 4 ; Shetelig, Cruciform Brooches ofNorwqy, 99 ; Aberg, op. cit., 13).The brooch stand,:n fact almost apart from Aberg's Group r. Its closest kin arethose from Kempston, Beds. (Antiq. Journ., XlII, 240 , pI. xxxv), one from Grave122 at Abingdon, Berks. (Oxoniensia, Vll, L02, pI. V]]], A; found after the mainexploration), and one from Grave 17 at Nassington, Northants. (Antiq. Joum.,XXIV, L07 and "7 , pI. xxv, a). The fibula is purely a north-Germanic type.(b) For the back-plate of he composite brooch, see remarks on the material fromGrave III.(c) Buckle. The nearest parallel known from this country is the specimenfound at Stanwick, Yorks. N.R . (B.M. Anglo-Saxon Guide, fig . LOS ; Leeds, EarlyAnglo-Saxon Art and Archaeology, 19, fig. 6) . The Stanwick buckle even morethan the Dorchester piece declares its origin in the pair of confronted birds, as


11/16

THREE EARLY SAXON GRAVES FROM DORCHESTER, OXON.represented on many Christian sarcophagi, etc., of the latest Roman days inGaul. The style, as has been shown, is well represented in other ways incemeteries of north-eastern France and Belgium. That this type of buckle hada ,vidish vogue in Britain is indicated by the find-spots of other examples \vithsimilar hoops, but lacking the plate. Such are known from Castor, Northants .(Artis collection, Peterborough Mus.), Bifrons, Kent (Baldwin Brown, i l l , pI.Lxx,6; Leeds, op. cit., pI. xm), and from a grave at Alveston, Wanvicks. Thetwo latter can certainly be regarded as collected from natives or native siteshy Anglo-Saxon invaders.GRAVE III

The pair ofcomposite brooches belong-as do the backplale from this grave andanother from Grave II, to a type which is to be found coming into fashion in thelater days of the Western Empire, and as shown by Roeder (Die siichsischeSchalenfibel der VOlkerwanderungszeit, Gottinger Beitrage zur deutschen Kulturgeschichte, 1927, p. 39) was widely accepted by the north German tribes. Theform is unquestionably the forerunner of the Anglo-Saxon' applied' brooch,in which protection was given to the fragile embossed disk by the addition of avertical rim brazed on to the back-plate, a development apparently unknownin north Germany.

There the normal decorative design is the five-fold spiral, common onlater English saucer-brooches, the north German prototype of which is knownfrom Caister by Norwich (Leeds, op . cit., 39, pI. XI , b). The ' komponierteFibel' of north Germany is possibly represented also in England by the broochwith spiral decoration from grave 106 at Abingdon, where it was associatedwith a damaged IUlulus-brooch, also ornamented with spiral-coils, probablyof earlier date than previously suggested.

The design employed on the Dorchester pair is again five-fold in character,but seems to be in no way related to the spiral. Roeder (op . cit., 8, pI. II, I) citesat the head of his remarks on this type a brooch from Friesland (PL. IV , c) ,found in a terp atJouswier, near Dokkum (P. C. J. A. Boeles, Friesland tol de eljdeEeuwe, 174, pI. xxxn, II ). Roeder was almost at a loss to interpret thedecorative design, which, curiously, he compared to that of the disk-attachments described under Grave 1. Actually close examination of Boeles's andRoeder's figures places beyond cavil its identity with that on the Dorchesterpair. The origin of the design is not evident, but in it we may possibly see theinspiration which, when the passion for zoomorphic ornament had obtained afirm hold on north Germanic art, produced that essentially Anglo-Saxon designof a whirl of legs, which m o ~ e o v e r frequently appears in a five-fold arrangement.

73


12/16

JOAN R. KIRK A D E. T. LEEDSApart from these, the back-plate of a third composite brooch, possibly thefinger-ring and the pricker, all the remaining relics from the grave are of

Roman origin. The heavier rings are never normally associated with AngloSaxon grave-furniture. The same applies to the bracelets with notched edges.They have been found in Saxon graves; as at Chatham Lines.

The brooches from Grave I I I alone are insufficient to establish a Friesianorigin for the occupants of the graves. The foreign (non-Roman) materialcan derive from any point between the Elbe and the Zuider Zee. At the sametime there are certain phenomena in the archaeology of Anglo-Saxondom thatoffer a strong presumption in favour of a considerable part of the early settlershaving come from north Holland rather than from farther east, e.g. the closeaffinity between much of the Anglo-Saxon pottery in Kent with that fromDutch sites.

CONCLUSIO TSIn the matter of the date of these graves and the origin of their occupants,the above comparisons render self-evident the existence of a very perceptible

gap between the contents of these three burials and the general mass of materialnormally assigned to the early invasion settlements. On the one hand we haveobjects markedly Romanized in character, some of which, like the sporran-top,have been frequently found in north-western Europe in a Germanic association,or objects purely Roman producing the impression that they had been collectedby the deceased in his or her lifetime (Grave III ), or a predominance ofGermanic along with a slighter sub-Germanic element (Grave II ). There isnothing that conforms to the pattern of the true invasion period; there is notypical Anglo-Saxon cemetery material, as at Long Wittenham or Abingdon.The non-appearance of such a cemetery at Dorchester is one of the puzzles ofupper Thames archaeology. These burials are reminiscent of what have beentermed' fighting' burials in Wiltshire and elsewhere, or, to take an examplenearer home, like those in Stanton Harcourt barrow (O xoniensia, x, 33 ff.), wherean ancient tumulus was selected as a pagan Saxon graveyard. In all thesepersists the tradition of the beorh picture in the Beowulfpoem. I t seems to havebeen the early practice adopted in the case of individuals or quite small groupsas distinct from the communal practice of the settlement period. For thepioneers, whether singly or in small numbers, cremation would be excluded.Precautionary reasons against announcing their presence would deny them theblaze of a funerary pyre.

The Dorchester graves are, however, not conditioned by any such precautions, since they can with full confidence be assigned to the pre-settlement74


13/16

THREE EARLY SAXON GRAVES FROM DORCHESTER, OXON.times, by which date Roman cremation was a dead letter. We have thus toseek another explanation. During the decline of Roman power in the westand indeed for long before that-natives living close to the limes and even frombeyond it were quite commonly enrolled or recruited into the Roman army, sothat it is only natural that in north-eastern France and Belgium there shouldhave developed a Romano-Germanic culture, tending to become more andmore Germanic as Roman power declined and the Frankish influx increased.It is these enlisted persons who can be held responsible for the frequent appearance of romanizing trappings and jewellery in north German burials or Germanic types in semi-Roman cemeteries. In short, both the date of theromanizing as of the unquestionably Germanic objects in the Dorchestergraves point not to invaders pure and simple, bu t to foederati in some Romandetachment stationed at Dorchester. 'Was part of their duty protection of the camp or refuge-area between theDyke Hills and the junction of the two rivers, Thames and Thame, overlookedby the older camp on Wittenham Clumps? Were the burials made in thebanks before the withdrawal of the legions? And had these foederati intermarried with British natives or had they brought their women-folk with them?The generally accepted dating of the relics makes either explanation possible.For the fibula in Grave II the last quarter of the fourth century is agreed, andthe continental cemeteries from which so many parallels have been drawn areassigned to a period between A.D. 350 and 450. Roeder also dates the development of his komponiertl Fibe/ as beginning in the last half of the fourth century.He points out that the cast saucer-brooch in Germany is a regional type, confined to the eastern moorland (Geest) of the governmental district of Stade.Strangely no example occurs from cemeteries like Wester-Wanna, Loxstedt, etc.By the time of the great descent on Britain the saucer-brooch, as illustratedby the material from Sussex graves, was in full production. The circularbrooches from Dorchester are two generations earlier.

The alternative identification of the persons buried in the Dyke Hillssuggested by Collingwood and Myres (Roman Britain and the English Settlements,pp. 394-5), is not very convincing. It is hardly probable either that a raidingparty or a nest of river-pirates would, in the closing year of Roman occupation,have brought womenfolk with them. But foederati transferred from theContinent to supplement or replace the steadily diminishing legionary forcescould well have been stationed at Dorchester to stem any southward thrust ofinvaders, and these foederati could have been accompanied by their women.The warrior had clearly been in contact with the late fourth century culture ofthe Vermand type, but the women must have been natives of north Germany

, We are indebted to Professor 1. A. Richmond for information about thefOtdnati.75


14/16

JOAN R. KIRK AND E. T. LEEDSor Friesland; they could hardly otherwise have acquired the brooches theywore.

Probably light could be thrown on the whole question by a thoroughinvestigation both of what still remains of the Dyke Hills and of the area theywere designed to protect. A mine-detector run along the tops of the banksmight well reveal more burials. Actually one, unfortunately without relicswas found in '940 in the top of the northern mound close to its middle poin t(Oxoniensia, v (1940), p. 163); another, also without relics was detected in '943by members of a nearby American camp in the bottom of a slit-trench in thetop of the western end of the northern bank at a depth of3 feet, and orientated,like those found in 1874, SSE. and so directly in the line with the axis of thebank itself.


15/16

PIAI E 1\

B (A. B. L)ORCI-IES I 1R. ()XON.

A. O setIon a r P 1p. . 7' t11 Sc;:l,B. Broolis and buCkIC rI#m (ira\e I pp. 67 IE I72 51ta 1(. FRII SLVND

ApplI d Inrof,( h I oti .1iissit-r. near Dtkkurtit 1p73 S lt%I] I I , I, I Mus ,A

OXONIENS[51 , V(OL X I I X\IllI ,1


16/16

BI}ORjI- IFSER, ()XC}N.

\. ( )Iects fromGrav II :pp.6c I. Sca' :13. X1 :plIcd }brc oche lcrn1 Grat{ 1IIi pp. 6!}. 73 lI.. Scol,:OXEIN iNSJ,I \ \ l 'ii. XVIII KIaRIKANIDIElisD. I}II{EF2 SAXON KRA\I:S

Kirk - Three Early Saxon Graves From Dorchester, Oxon

Documents

Transcript of Kirk - Three Early Saxon Graves From Dorchester, Oxon